Reinforcement Learning on Atari 56 tasks benchmark leaderboard · SOTA2 Research

Reinforcement Learning on Atari 56 tasks

31.33Mean ScoreRL-D2

Evaluation Results

Method	Links
RL-D22025.09		31.33	4.09	4.98
IMPALA-Macro2025.09		18.01	2.36	2.69
IMPALA2025.09		16.55	2.33	3.17
DQN2025.09		9.49	1.77	1.89
DQN-Macro2025.09		7.54	1.51	1.91
R2D22025.09		6.31	1.32	1.55
PPO2025.09		4.47	1.73	1.9